Analyzing recommender systems and applying a location based approach using tagging

In the past few decades, due to the increase in the accessibility of the world wide web, data is being generated on a large scale, everyday. When the end user tries to access this data, the available variety of options make it a difficult task to choose amongst various products. In order to simplify this process, Recommender Systems were designed. In this paper, we have analyzed how these Recommender Systems came into existence, their functionalities and how this idea is shaping the world of data science. We analyzed few of the prominent techniques used, such as Content based, Collaborative and Hybrid recommendations. Tagging has become an important tool in Data Science, which can be used to collect implicit data of users and generate recommendations. The paper proposes such an approach which is motivated by observing that many effective recommendations can be generated by analyzing the tagging done by the users. Most of the work is based on the feedback from users having a tendency to use the tagging feature, mostly on social media platforms. The tag based recommendations are also generated by taking into consideration a pattern between users who use location-based services. Counting in the similarities between their Internet browsing preferences and the tag information collected, accurate recommendations can be suggested to the users.